## Unleashing AI Power: Introducing the All-New Lenovo ThinkStation PGX
Lenovo has just rolled out **the ThinkStation PGX**, a powerful new personal AI developer workstation that’s *revolutionizing the way AI researchers, developers, and students interact with cutting-edge technologies*. This compact yet robust solution is designed for those who require a **purpose-built desktop** that’s ready to deploy right out of the box.
### **Exceptional Performance with NVIDIA Superchip**
At the heart of the ThinkStation PGX lies the **NVIDIA GB10 Grace Blackwell Superchip**, delivering an astounding **1 PetaFlop** (or 1000 TOPS)[^1]. This powerhouse enables users to tackle generative AI models with up to **200 billion parameters**. With **128 GB of coherent unified system memory**[^2], developers have the freedom to experiment, fine-tune, or infer the latest generation of reasoning AI models.
#### **Double the Power, Double the Creativity**
For those looking to push the envelope even further, two ThinkStation PGX units can be seamlessly connected. This ingenious setup allows them to manage **AI models up to 405 billion parameters**, significantly broadening the horizons for innovation and creativity.
### **Streamlined AI Development Environment**
The ThinkStation PGX arrives preconfigured with the **NVIDIA DGX OS** and the comprehensive **NVIDIA AI software stack**. Integrated with familiar tools like **PyTorch** and **Jupyter**, it empowers developers to **instantly prototype**, fine-tune, and deploy large AI models directly from their desktops.
> “By collaborating with NVIDIA to deliver a high-performance, yet compact device, Lenovo is empowering AI developers, researchers, data scientists, and students to accelerate their workloads and adoption of breakthrough innovation in generative AI.”
> **— Rob Herman, Vice President, Worldwide Workstation and Client AI Business at Lenovo**
### **Facing the Challenges of Generative AI**
As the size and complexity of generative AI models continue to expand, developing them on traditional local systems becomes increasingly challenging. Prototyping and tuning these models often require substantial **GPU memory** and **performance**. The ThinkStation PGX stands out by offering a powerful yet economical platform, enabling developers to optimize their productivity while freeing up essential resources for larger-scale training and deployment.
### **Seamless Transition to Cloud Infrastructure**
One of the most significant benefits of the ThinkStation PGX is its ability to leverage the **NVIDIA AI platform software architecture**. This integration allows users to transition their models effortlessly from desktop to any accelerated **cloud or data center infrastructure**—all with minimal code changes. The result? An unparalleled ease of prototyping, fine-tuning, and iteration.
### **Lenovo’s Commitment to AI Innovation**
Lenovo’s dedication to delivering AI-ready performance extends across its entire workstation lineup, from powerful [ThinkStation desktops](https://techtoday.lenovo.com/ww/en/workstations) to the nimble ThinkPad P Series mobile workstations. Through **Lenovo’s Hybrid AI Advantage with NVIDIA**, enterprises gain access to full-stack, validated AI solutions, making it easier to develop and deploy **agentic AI**—systems capable of efficient reasoning, planning, and action. These scalable solutions allow businesses to access AI across various environments, including **edge computing** and **cloud infrastructures**, thus empowering productivity and driving innovation.
### **Availability and More Information**
The ThinkStation PGX is set to hit the market in **calendar Q3 of 2025**. For those intrigued by this groundbreaking workstation, more details can be found on [Lenovo’s official webpage](http://www.lenovo.com/us/en/workstation-ai).
—
[1] 1 petaflop of FP4 AI precision using sparsity
[2] Available user storage and internal memory is less due to the operating system, software, and other functions utilizing part of this capacity; may change with software updates.
